Estimate total trainable parameters and forward-pass FLOPs for a fully-connected neural network.
Estimates total trainable parameters and forward-pass FLOPs for a fully-connected (dense) network from its layer sizes.
| Layer | Shape | Parameters | FLOPs (per sample) |
|---|---|---|---|
| 1 | 784 → 256 | 200,960 | 401,664 |
| 2 | 256 → 128 | 32,896 | 65,664 |
| 3 | 128 → 10 | 1,290 | 2,570 |
FLOPs counted as 2 per multiply-add (industry convention). Only covers fully-connected layers — convolutional/attention layers have different FLOP formulas not modeled here.
Compute PDF/PMF and CDF values for the normal, binomial, and Poisson distributions.
Compute accuracy, precision, recall, specificity, and F1 score from a binary confusion matrix.
Compute mean, median, mode, variance, standard deviation, and quartiles from a list of numbers.
Run a one-sample t-test, two-sample (Welch's) t-test, or chi-square test and get the test statistic and p-value.
Solve a system of up to 5 linear equations (Ax = b) using Gaussian elimination.
Calculate the minimum sample size needed to estimate a proportion within a chosen margin of error and confidence level.
Numerically evaluate a definite integral of a function between two bounds using Simpson's rule.
Compute the symbolic derivative of a single-variable function and evaluate it at any point.
Computes total trainable parameters (weights + biases) and estimated forward-pass FLOPs for a fully-connected (dense) network, given only its layer sizes. FLOPs are counted as 2 per multiply-add, the standard convention used in most model efficiency reporting.
Covers dense/linear layers only — convolutional and attention layers have different FLOP formulas not modeled by this calculator.
Did you find this tool helpful? Tell us how we can make it even better, or report a bug or request a new feature in under 1 minute.